Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. A method for replacing missing samples in a signal comprising a plurality of discrete samples, comprising: receiving a plurality of the discrete samples, each discrete sample having a respective sample location in a set of sample locations; identifying a subset of sample locations representing missing samples in the signal; determining a first threshold and a second threshold, wherein the first and second thresholds are each an integer number of samples, and the second threshold is greater than the first threshold; forming a first set of consecutive sample locations from the identified subset of sample locations; and replacing the missing samples in the first set of consecutive sample locations using a replacement process selected based at least in part on a comparison between a number of locations in the first set of consecutive locations, the first threshold, and the second threshold, and wherein the selected replacement process comprises signal modeling if the number of locations is between the first and second thresholds.
A method for fixing missing data points in a signal (like sensor readings or network data) involves these steps: First, receive a set of data points, each with a timestamp or location. Then, identify which data points are missing. Define two thresholds (first and second), representing small and large gaps of missing data points respectively. When a continuous block of missing data is found, the method selects how to fill the gap by comparing the gap size to the two thresholds. If the gap size falls between the two thresholds, signal modeling is used to estimate the missing data based on trends or patterns found in the existing data.
2. The method of claim 1 , wherein the samples in the signal correspond to a number of data packets arriving at a location in a network within a time interval.
The method described above for fixing missing data points, where the data represents network packets arriving at a specific location within a set period of time. This means that the "signal" is the stream of network packets, and missing samples represent lost or delayed packets. The method fills in these missing packet arrival events.
3. The method of claim 1 , wherein the selected replacement process comprises signal interpolation if the number of locations is less than the first threshold.
The method for fixing missing data points, when the count of consecutive missing data points is below the first threshold, uses signal interpolation to estimate their values. Signal interpolation means estimating missing values based on the values of the data points immediately before and after the missing section. This is suitable for small gaps in the data.
4. The method of claim 1 , wherein signal modeling comprises identifying a trend component of the signal based on a plurality of received samples.
In the method for fixing missing data points, the "signal modeling" process includes identifying a trend component within the received data. This means that the method analyzes the existing data points to find any increasing, decreasing, or other directional patterns. This trend is then used to estimate the values of the missing data points.
5. The method of claim 1 , wherein signal modeling comprises identifying a cyclical component of the signal based on a plurality of received samples.
In the method for fixing missing data points, the "signal modeling" process involves detecting any recurring, cyclical patterns in the available data points. This means finding patterns that repeat over time, such as daily or weekly fluctuations in the data values. These identified cyclical components can improve the accuracy of estimating missing values.
6. The method of claim 5 , further comprising determining a period of the cyclical component of the signal, wherein: the first threshold is determined to be at least an order of magnitude less than the determined period; and the second threshold is determined to be on a same order as the determined period.
The method for fixing missing data points, when a cyclical component is found, the method determines the cycle's duration. The first threshold (small gap) is set to be much smaller than this cycle duration. The second threshold (large gap) is set to be approximately the same length as the cycle duration. This ensures that short gaps are handled differently than longer gaps relative to the signal's inherent periodicity.
7. The method of claim 1 , further comprising receiving a plurality of quality values, wherein each quality value in the plurality of quality values is an amount of accuracy of a corresponding discrete sample.
The method for fixing missing data points also includes receiving a set of "quality values," where each value represents the accuracy or reliability of a corresponding data point. This adds a measure of confidence to each data point, helping the system decide whether a sample is truly "missing" or just unreliable.
8. The method of claim 7 , wherein the subset of sample locations representing missing samples in the signal is identified by identifying discrete samples with corresponding quality values below a quality threshold.
In the method for fixing missing data points that uses quality values, a data point is identified as "missing" if its corresponding quality value falls below a predefined quality threshold. This allows the system to automatically detect unreliable samples based on their quality metrics.
9. The method of claim 7 , wherein the replacement process is based at least in part on a plurality of quality values corresponding to a plurality of identified non-missing samples.
In the method for fixing missing data points that uses quality values, the data replacement process (interpolation, signal modeling) uses the quality values of non-missing data points to improve the accuracy of the missing data estimation. This could involve weighting nearby data points based on their quality when performing interpolation or trend analysis.
10. A computer apparatus for replacing missing samples in a signal comprising a plurality of discrete samples, comprising: a memory containing instructions that, when executed by one or more processors, cause the one or more processors to perform a set of steps comprising: receiving a plurality of the discrete samples, each discrete sample having a respective sample location in a set of sample locations; identifying a subset of sample locations representing missing samples in the signal; determining a first threshold and a second threshold, wherein the first and second thresholds are each an integer number of samples, and the second threshold is greater than the first threshold; forming a first set of consecutive sample locations from the identified subset of sample locations; and replacing the missing samples in the first set of consecutive sample locations using a replacement process selected based at least in part on a comparison between a number of locations in the first set of consecutive locations, the first threshold, and the second threshold, and wherein the selected replacement process comprises signal modeling if the number of locations is between the first and second thresholds.
A computer system designed to fix missing data points in a signal (like sensor readings or network data) using programmed instructions. The system receives a set of data points, each with a timestamp or location and identifies which points are missing. It defines two thresholds, representing small and large gaps. For any consecutive block of missing points, the system compares the block size to the two thresholds. Based on this comparison, the system selects a way to fill the gap. If the gap size falls between the two thresholds, the system uses signal modeling to estimate the missing data by analyzing the existing data for patterns.
11. The apparatus of claim 10 , wherein the samples in the signal correspond to a number of data packets arriving at a location in a network within a time interval.
The computer system described above for fixing missing data points, where the data represents network packets arriving at a specific location over time. The "signal" is the stream of network packets, and missing data points signify lost or delayed packets. The system aims to reconstruct the missing packet arrival events.
12. The apparatus of claim 10 , wherein the selected replacement process comprises signal interpolation if the number of locations is less than the first threshold.
In the computer system for fixing missing data points, when the count of consecutive missing data points is below the first threshold, the system uses signal interpolation to estimate their values. Signal interpolation estimates missing values based on the data points immediately before and after the missing section.
13. The apparatus of claim 10 , wherein signal modeling comprises identifying a trend component of the signal based on a plurality of received samples.
In the computer system for fixing missing data points, the "signal modeling" process includes identifying a trend component within the received data. The system analyzes the data to find any directional patterns, and it uses this trend to estimate the missing data points.
14. The apparatus of claim 10 , wherein signal modeling comprises identifying a cyclical component of the signal based on a plurality of received samples.
In the computer system for fixing missing data points, the "signal modeling" process includes detecting any recurring, cyclical patterns in the available data. The system identifies patterns that repeat over time, such as daily or weekly data fluctuations to estimate the missing values.
15. The apparatus of claim 14 , further comprising a processor for determining a period of the cyclical component of the signal, wherein: the first threshold is determined to be at least an order of magnitude less than the determined period; and the second threshold is determined to be on a same order as the determined period.
The computer system for fixing missing data points where the data includes cyclical patterns, the system determines the cycle's length. The first threshold (small gap) is set much smaller than the cycle duration, and the second threshold (large gap) is set approximately equal to the cycle duration. This ensures that small gaps are handled differently than larger gaps when compared to the signal's inherent rhythm.
16. The apparatus of claim 10 , further comprising a processor for receiving a plurality of quality values, wherein each quality value in the plurality of quality values is an amount of accuracy of a corresponding discrete sample.
The computer system for fixing missing data points also receives "quality values" for each data point, representing the data's accuracy or reliability. This added information helps the system determine if a data point is truly missing or just unreliable.
17. The apparatus of claim 16 , wherein the subset of sample locations representing missing samples in the signal is identified by identifying discrete samples with corresponding quality values below a quality threshold.
The computer system for fixing missing data points that uses quality values identifies a data point as "missing" if its corresponding quality value is below a predefined quality threshold. This allows the system to automatically flag unreliable samples based on their quality metrics.
18. The apparatus of claim 16 , wherein the replacement process is based at least in part on a plurality of quality values corresponding to a plurality of identified non-missing samples.
The computer system for fixing missing data points that uses quality values utilizes the quality values of existing, non-missing data points during the replacement process (interpolation or signal modeling). This means it incorporates the confidence level of nearby data when estimating missing values.
Unknown
October 21, 2014
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.